EVALUATION OF THE SUCCESS OF CONSTRUCTED WETLANDS IN THE CAVE RUN LAKE WATERSHED A Thesis Presented to the Faculty of the College of Science and Technology Morehead State University In Partial Fulfillment of the Requirements for the Degree Master of Science by April Diane Haight December 8, 1996 Ay>f'- f'"\ T\-.. e :5 • -o l,:39,9 2- \-\ l l\ 5 .ll..- Accepted by the Faculty of the College of Science and Technology, Morehead State University, in partial fulfillment of the requirements of the Master of Science degree. 3--:::-:=::: (.: *CC"), Director of Thesis Master's Committee: ii EVALUATION OF THE SUCCESS OF CONSTRUCTED WETLANDS IN THE CAVE RUN LAKE WATERSHED April D. Haight, M.S. Morehead State University, 1996 Director of Thesis: ~ (;,ye 9------.. ABSTRACT Beaver Creek Wetland Complex, a United States Forest Service wetland restoration project, gives a unique opportunity to study wetland development in Eastern Kentucky. We assessed five constructed wetlands (ranging from 5 to 3 years in age) and determined if avifuana and plant communities were similar to a nearby natural wetland. All of the constructed wetlands were attracting avifauna and plants dependent upon shallow water and/or saturated soil communities for their existence (obligate wetland-species). However, neither the plants nor birds were similar to the control site. For avifauna, community similarity to the control ranged from 0.243 to 0.635; for macrophytes community similarity was even lower, ranging from 0.116 to 0.186. Avifauna richness was significantly correlated to habitat diversity (r=0.792; p=0.06). All of the constructed wetlands had a greater macrophyte richness than the control. A sediment core of the control site allowed a look iii at the past communities development. The pollen stratigraphy indicated that the control site has been a wetland for the last 1,840 years. We found that the control site did not accumulate significant organic material (up to 28% organic matter) until after the nearby river was dammed. Deforestation seemed to alter the hydrology and allowed the establishment of Sphagnum. No significant correlations were found between organic matter and total phosphorus (r=0.316), total iron (r=0.079), or total sulfur (r=0.055) in the core sediments. Natural wetlands appear to be stable components of the Cave Run Lake watershed; constructed wetlands are currently not able to maintain a similar stability. Accepted by: -~-=----':;__C::...._~-==--------• Chair iv ACKNOWLEDGMENTS I would like to thank Dr. Jerry Howell for the many revisions of my writing; Dr. Howard Setser and Dr. Leslie Meade for helping with plant identification; Fred Busroe for advice concerning avifauna; Tom Biebighauser and Frank Bodkin of the USFS Morehead Ranger Station, for freely giving much pertinent information and Steve Davis and Tim Spier for help in the field. I would like to especially thank Dr. Brian Reeder for reminding me and everyone else that I had a Thesis to complete. Lastly, a special thanks to Dr. Walter Kealey for encouragement 13 years ago and my husband, John, for helping me through. V TABLE OF CONTENTS CHAPTER PAGE I. A Survey of Avifauna 1 1.1. Introduction 1 1.2. Literature Review 2 1.3. Site Description 5 1.4. Methods 9 1.5. Results 11 1.6. Discussion 20 1.7. Literature Cited in Chapter I 22 II. A Survey of Macrophytes and Soil Development 27 2.1. Introduction 27 2.2. Literature Review 27 2.3. Site Description 31 2.4. Methods 33 2.5. Results 34 2.6. Discussion 41 2.7. Literature Cited in Chapter II 44 Ill. Pollen and Soil Analysis of the Control Wetland 48 3.1. Introduction 48 3.2. Literature Review 49 3.3. Site Description 50 3.4. Methods 52 3.5. Results 54 3.6. Discussion 61 3.7. Literature Cited in Chapter Ill 65 IV. Conclusion and Recommendations 68 vi LIST OF TABLES , PAGE Table I. Basic Wetland Characteristics during ' ' ' Study Period 8 Table II. Summary of Avian Observations 12 Table Ill. Avifauna Community Similarity to Control Site 15 Table IV. Wetland Characteristics for Macrophyte Survey 32 Table V. Taxonomic Summary for Beaver Creek Wetlands and Control 36 Table VI. Macrophyte Richness and Community Similarity to Control Site 37 Table VII. Surface Sediment Analysis Summary (ug/kg) 40 vii LIST OF FIGURES PAGE Figure 1. Location of Wetland Complex and Natural Wetland 6 Figure 2. Percentage of Habitat by Wetland Site 13 Figure 3. Relationship between Habitat Diversity and Avifauna Species Richness of Wetland Studied. 18 Figure 4. Habitat Diversity and Avifuana Diversity of Wetlands Studied 20 Figure 5. Plant Category by Percentage for Each Wetland 38 Figure 6. Percent Organic Matter and Concentration of Nutrients (TP, Fe, TS) 55 Figure 7. Groups of Pollen Taxa and Depth • · 57 Figure 8. Pollen Percentage Diagrams of the Main Tree Taxa of the Sediment Core 58 Figure 9. Pollen Percentage Diagrams of the Main other Herbaceous Taxa of the Sediment Core 59 Figure 10. Pollen Percentage Diagrams of the Hydrophyte Taxa of the Sediment Core 60 Figure 11. Time Line of Events for the Area Surrounding the Control Site 64 viii Appendices APPENDIX PAGE A Morphology Maps 71 B Avifauna Observations by Wetland Site 77 C Plant Collection Data 89 D Soil Analysis 98 E Sediment Core Organic Matter and Bulk Density 100 F Sediment Core Nutrient Analysis 106 G Pollen Identification and Enumeration 110 ix CHAPTER! A Survey of Avifauna 1.1. Introduction Eighty percent of America's breeding bird population and more than 50 percent of the 800 species of protected migratory birds rely on wetlands (Wharton et al. 1982). Wetlands are the rarest habitat in the Daniel Boone National Forest (DBNF) (Beibighauser 1993); therefore, fifty wetlands · covering an area of 24 hectares (ha) have been constructed since 1989 at Morehead Ranger District in the DBNF to encourage the establishment of wetland habitat, and hence, waterfowl populations in the National Forest.. Waterfowl also act as an indicator of wetland health. If they are present, aquatic insects and hydrophytes that waterfowl are dependent upon will also be present. This research seeks to determine if the constructed wetlands are attracting obligate species and have bird populations established in the same richness, diversity and abundance as natural wetlands. 1.2. Literature Review 1.2.1. Hydrology Wetland hydrology and waterfowl use have a positive correlation (Weller 1994; Beauchamp et al. 1996). Catfish ponds in Mississippi act as artificial wetlands to waterfowl in a region decreasing in natural wetlands, and as a water source during drought periods when natural wetlands dry out (Christopher et al. 1988). Bethke and Nudds (1995) have found that, since 1 1951, drought, not agriculture, has been the reason for waterfowl decline in Canadian prairie-park lands. Batt et al. (1989) also found that hydrology had an effect on waterfowl breeding success in the prairie pothole region. Wetlands constructed to attract breeding and migratory waterfowl tend to have an average depth of about one meter. Wetlands managed for hunting tend to be mostly shallow water (<1 m) (Payne 1992), while wetlands managed for refuge have a greater ratio of deep water and shallow water (Boekhout et al. 1989). Logan (1975) and Slimack (1975) both suggested that impoundments designed for the American Black Duck (Anas rubripes) have 25-50% of the surface area less than 0.9 m deep. Lewis and Nelson (1988) also found a direct positive correlation between shallow water ~ 1 m deep) and American Black Duck populations. However, water levels of 0.45- 1.2 meters tend to establish a 50:50 ratio of open water to emergent vegetation -- considered optimal for duck populations (Farmes 1985). Ponds used to attract diving ducks tend to be deeper, and therefore have less emergent vegetation (Linde 1985; Lokemoen et al. 1994). Water levels > 1.5 m tend to inhibit plants that can attract breeding ducks (Payne 1992). Breeding waterfowl, such as Canada Geese (Branta canadensis), have been shown to prefer less than 10% emergent cover (Ringel man 1990). The best wetland complex should_ incorporate all water levels (moist-soil , shallowwater and open-water); however, specific management goals determine what areas will dominate (Payne 1992). Surface area is also an important factor in waterfowl use. Impoundments greater than 10 ha can be heavily used, especially by duck populations (Bates et al. 1988). Craig and Beal (1992) found that species 2 richness of wetland birds correlates to total marsh area, while Grover and Baldassarre (1995) found this untrue for beaver ponds. Larger wetlands support a greater assemblage of plants and animals by retaining water during droughts (McKinstry and Anderson 1994). Often, there is a lag time, after droughts, of recovery for waterfowl populations (Wiens 1977). 1.2.2. Management Techniques Most construction criteria call for hydrologic control structures to be placed in waterfowl-managed wetlands (Copelin 1961; Gough 1988; Carroll 1990). Draw-downs allow the manager to increase bird food supply by stimulating seed plant and benthic invertebrate growth (Kusler and Kentula 1990; Wilcox and Meeker 1992). During reproductive periods, waterfowl, such as Canada Geese, eat invertebrates to supplement protein needs for egg production (Prichert 1991 ). Draw-downs are commonly used to enhance emergent vegetation for waterfowl use (Kadlec 1962; Payne 1992). The response of plants to water level manipulation depends on the timing of annual draw-downs and the stage of succession in the impoundment; for example, smartweed (Polygonum spp.) seed production is enhanced when the wetland is drawn­ down to moist soil conditions in the spring (Payne 1992). Wetland managers draw-down in the early summer to establish emergent vegetation growth and then raise water levels to give the broods more open water in the fall.